This project revolves around data wrangling and analyzing our dataset. Firstly, we import all the required Libraries for our analysis. This was followed by gathering the necessary datasets that are required for our analysis. Overall, we analyzed three (3) dataset for this project and load them in the notebook through different approaches.
-
The first dataset which is the ‘the twitter-archive-enhanced’ dataset which was downloaded directly. We proceeded by reading it into a dataframe in other to perform both visual and programmatic assessment with some issue being detected and noted.
-
I then acquired the tweet image prediction dataset using the request library. Visual and programmatic assessment were also performed to understand the dataset to understand the various issues that are associated with it.
-
The Tweepy library was also used to query additional data via the twitter API provided in my Udacity Dashboard. since I was having challenges signing up for a Twitter Developer Account. This code took over 30 minutes to run.
Eventually, I succeeded in downloading the JSON Data. Visual and programmatic assessment were also appliedto the dataset to understand both the quality and tidiness issues as well as the correlation between all the ttributes inherent in them. At the end of this first stage, I was able to itemize a total of nine (9) quality and two (2) tidiness issues in the three (3) datasets given.
This repository contains the following files:
- An HTML version of my final Jupyter Notebook;
- The ipynb Jupyter Notebook file for executing my Python scripts
- The CSV file that contains the project dataset.
- Twitter API
For any questions or comments about this repository, please feel free to email me at [email protected] or on Twitter at @breeze099.